Skip to content

Latest commit

 

History

History

ocrnet

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
README.md
README.md
 
 

README.md

Introduction

Official Repo

Code Snippet

OCRNet (ECCV'2020) 
@article{yuan2019object,
    title={Object-contextual representations for semantic segmentation},
    author={Yuan, Yuhui and Chen, Xilin and Wang, Jingdong},
    journal={arXiv preprint arXiv:1909.11065},
    year={2019}
}

Results

PASCAL VOC

Backbone Pretrain Crop Size Schedule Train/Eval Set mIoU Download
R-50-D8 ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 76.75% cfg | model | log
R-101-D8 ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 78.82% cfg | model | log
HRNetV2p-W18-Small ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 72.80% cfg | model | log
HRNetV2p-W18 ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 75.80% cfg | model | log
HRNetV2p-W48 ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/60 trainaug/val 77.60% cfg | model | log

ADE20k

Backbone Pretrain Crop Size Schedule Train/Eval Set mIoU Download
R-50-D8 ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 42.47% cfg | model | log
R-101-D8 ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 43.99% cfg | model | log
HRNetV2p-W18-Small ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 38.04% cfg | model | log
HRNetV2p-W18 ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 39.85% cfg | model | log
HRNetV2p-W48 ImageNet-1k-224x224 512x512 LR/POLICY/BS/EPOCH: 0.01/poly/16/130 train/val 44.03% cfg | model | log

CityScapes

Backbone Pretrain Crop Size Schedule Train/Eval Set mIoU Download
R-50-D8 ImageNet-1k-224x224 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/440 train/val 79.40% cfg | model | log
R-101-D8 ImageNet-1k-224x224 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/440 train/val 80.61% cfg | model | log
HRNetV2p-W18-Small ImageNet-1k-224x224 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/440 train/val 79.30% cfg | model | log
HRNetV2p-W18 ImageNet-1k-224x224 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/440 train/val 80.58% cfg | model | log
HRNetV2p-W48 ImageNet-1k-224x224 512x1024 LR/POLICY/BS/EPOCH: 0.01/poly/8/440 train/val 81.44% cfg | model | log

More

You can also download the model weights from following sources: